INDEX
Explanations
the pronoun "it" in various contexts throughout the text
New Auto-Interp
Negative Logits
Dollar
-0.68
idth
-0.63
herent
-0.63
ãĥ©ãĥ³
-0.60
arthed
-0.58
owed
-0.58
hips
-0.57
Monetary
-0.57
quartered
-0.56
Unified
-0.56
POSITIVE LOGITS
alian
1.06
beh
0.98
unes
0.96
depends
0.95
relates
0.94
chy
0.94
boils
0.94
hurts
0.92
'll
0.91
's
0.89
Activations Density 0.090%